William Withering's legacy--for the good of the patient.

The lessons that the physician William Withering learned from his studies of digitalis are still relevant today. This paper highlights four of these lessons and updates them using the tools of clinical pharmacology and pharmacoepidemiology. First, Withering learned that failure to prepare digitalis from the foxglove in a standard manner resulted in a product with unpredictable clinical effects. Preparation of medicines from plants since then has not followed similar good practice and medicines have often not been granted marketing authorisation because of variability in their quality. Second, differences in the response to digitalis were noted by Withering, but he had little idea of their basis. Clinical pharmacology has shown that for drugs such as digitalis differences are caused by variability both in receptor sensitivity and in drug disposition. Third, the dose-response characteristics of digitalis were well known to Withering. Modern techniques of measuring response, such as the use of biomarkers, have made such studies easier, although clinical observations remain the gold standard. Fourth, Withering documented many of the adverse effects of digitalis. The use of various modern databases has facilitated the analysis of clinical toxicology and thus of risk-benefit profiles.